Low-Latency Analog Processors for Performance Control of MEMS

Y. Han, Z. Zhu, A. Abrol, J. Clark Auburn University, United States

Keywords: PODMEMS, feedback electronics, BJT

Summary:

For low-latency signal processing of feedback for controlling the response behavior of MEMS, we explore the use of new analog processing circuits based on bipolar junction transistors (BJTs). By monitoring the state of the proof mass of a MEMS device, and feeding back a force onto the proof mass that is proportional to (or a function of) the state, then the apparent mass, damping, and stiffness of the MEMS device can be significantly altered. To maximize stability, it is necessary that the sensed state be processed and fed back in the smallest amount of time. Typically, tens of nanoseconds are a low enough latency for MEMS that operate on a time-scale of tens to hundreds of kilohertz. A larger domain of stability allows a MEMS device to achieve a larger range of values for apparent mass, damping, and stiffness. Our prior work studied the effect of using analog feedback circuit comprised of filters and operational amplifiers, which resulted in latency of fifty nanoseconds [1]. The latency of the present work is at least five times smaller. We model the MEMS and BJT-based feedback electronics as an equivalent circuit, which we simulate using a circuit solver. Due to high input impedance, small nonlinear distortion, high-temperature stability and a good ability of anti-interference of those circuits, extremely small currents from the state sensors of the MEMS can be detected and shaped. Meanwhile, the feedback circuit used has a high operating speed and a small delay, which will improve the stability of the whole device by a great amount and greatly extend the dynamic performance range. The paper will also present characterizations of our BJT-based processing components, such as addition, multiplication, squaring, cubing, square rooting, integration, and differentiation. These circuit components will be combined to perform the necessary mathematical operations on the feedthrough signal of the MEMS device to achieve the desired response behavior. Our analyses will also include stability, sensitivity, electronic noise, and their effects on the control of MEMS behavior. [1] Clark J V, Misiats O, Sayed S. “Electrical control of effective mass, damping, and stiffness of MEMS devices”. IEEE Sensors Journal, 2017, 17(5): 1363-1372.